Hacker News front pageSep 25, 2026, 04:28 UTC+8

LaunchVideo turns a URL or prompt into an explainer video with Opus 5.5 and a headless renderer

Opus 5.5 is good at explainer videos

LaunchVideo generates a ~30-second product explainer from a URL or a text prompt. Opus 5.5 writes the HTML/CSS/animation script, and a serverless agent renders it frame by frame in a headless Chromium microVM — no video generation model is used. Each video costs roughly 100k tokens and takes about four minutes, outputting 1080p 30fps MP4 with a virtual clock for deterministic frames. The page shows five unedited examples including NVIDIA and Linear. The whole product is one TypeScript agent file plus three tools, fully open-source and one-click deployable to your own OpenComputer account. The post doesn't mention pricing or whether models other than Opus 5.5 are supported.

Why it matters: A clever packaging of Opus 5.5's coding ability into a 'URL-to-launch-video' tool, with a clearly explained pipeline and visible examples. But the product is still lightweight—more a sharp demo than an industry-shaking release. H and K both hit, R is weak, landing right at the...

Read the original ↗Export Markdown

Take

This caught my eye because it takes a completely different path to AI-generated video. No diffusion models, no Sora — Opus 5.5 writes the HTML, CSS, and animation scripts directly, then a headless Chromium microVM renders it frame by frame into a 1080p 30fps MP4. Each video burns roughly 100k tokens and takes about four minutes. The page shows five unedited examples including NVIDIA and Linear, and they look cleaner than I expected.

The whole thing is one TypeScript agent file plus three tools, fully open-source, deployable to your own OpenComputer account in one click. The rendering trick is clever: they replace all the page clocks — requestAnimationFrame, Date, CSS animations — with a virtual clock so every frame is deterministic. No video model means no hallucinated frames or consistency issues.

Two discounts I'd apply. One, the post doesn't mention pricing, and 100k tokens through Opus 5.5 isn't cheap — you'll want to run your own math. Two, it's locked to Opus 5.5 right now with no mention of other model support, so flexibility is limited. But the core idea is solid: reframe video generation as "code writes the animation, browser records it," sidestepping the quality and consistency headaches of video models entirely. If you've got a product landing page and want a quick 30-second explainer, this is way cheaper than hiring a motion designer.

Source: Hacker News front page · launchvideo.io
Published: Sep 25, 2026, 04:28 UTC+8 · 5 days ago